The combination of epigallocatechin gallate and curcumin suppresses ERα-breast cancer cell growth in vitro and in vivo

Abstract

Both epigallocatechin gallate (EGCG) and curcumin have shown efficacy in various in vivo and in vitro models of cancer. This study was designed to determine the efficacy of these naturally derived polyphenolic compounds in vitro and in vivo, when given in combination. Studies in MDA-MB-231 cells demonstrated that EGCG + curcumin was synergistically cytotoxic and that this correlated with G₂/M-phase cell cycle arrest. After 12 hr, EGCG (25 μM) + curcumin (3 μM) increased the proportion of cells in G₂/M-phase to 263 ± 16% of control and this correlated with a 50 ± 4% decrease in cell number compared to control. To determine if this in vitro result would translate in vivo, athymic nude female mice were implanted with MDA-MB-231 cells and treated with curcumin (200 mg/kg/day, po), EGCG (25 mg/kg/day, ip), EGCG + curcumin, or vehicle control (5 ml/kg/day, po) for 10 weeks. Tumor volume in the EGCG + curcumin treated mice decreased 49% compared to vehicle control mice (p < 0.05), which correlated with a 78 ± 6% decrease in levels of VEGFR-1 protein expression in the tumors. Curcumin treatment significantly decreased tumor protein levels of EGFR and Akt, however the expression of these proteins was not further decreased following combination treatment. Therefore, these results demonstrate that the combination of EGCG and curcumin is efficacious in both in vitro and in vivo models of ERα- breast cancer and that regulation of VEGFR-1 may play a key role in this effect. © 2007 Wiley-Liss, Inc.

In recent years, certain naturally occurring compounds have been recognized as potential treatments for breast cancer. Two of these compounds include epigallocatechin gallate (EGCG) and curcumin, both of which are plant-derived polyphenols. Specifically, EGCG is a flavonoid found in high levels in green tea1 whereas curcumin is a phytochemical derived from the rhizome of turmeric (Curcuma longa Linn), a spice used as a coloring and flavoring agent in Indian cooking.2 Epidemiological evidence has suggested that the consumption of foods containing high levels of either EGCG or curcumin may decrease the risk of developing numerous types of cancer, including breast cancer.3–6 For example, increased green tea consumption has been associated with both improved prognosis and decreased recurrence rate following the diagnosis of breast cancer.7

Results from experimental studies have supported the theory that dietary consumption of EGCG and curcumin may decrease cancer incidence. Both EGCG and curcumin inhibit the growth of a variety of human cancer cells lines in vitro, including estrogen receptor negative (ERα-) human breast cancer cells.8–12 The cytotoxic effect of both EGCG8, 13–16 and curcumin17–20 has been attributed to the induction of cell cycle arrest in the G₁/G₀ and G₂/M-phase, respectively, and subsequent apoptosis induction. Likely mechanisms for this effect include inhibition of the transcription factor nuclear factor-κB (NF-κB)21 and consequently various NF-κB-regulated gene products,22–24 as well as inhibition of cyclins and cyclin dependent kinases.8, 25 Furthermore, numerous in vivo studies have demonstrated a decrease in tumor initiation and growth following curcumin26–31 or EGCG administration.8, 32–36 For example, EGCG and green tea extracts were both recently shown to decrease tumor incidence and burden in athymic mice inoculated with MDA-MB-231 (ERα-) human breast cancer cells.8, 36 A potential mechanism for this effect is via inhibition of the epidermal growth factor receptor (EGFR), which is overexpressed in ERα-breast cancer cells37 and patients.38 Importantly, EGCG inhibited the activation of the EGFR in MDA-MB-231 breast carcinoma cells,39 indicating that this may also occur in in vivo xenograft models using this cell line.

Importantly, some naturally occurring compounds with anticancer properties are known to act in synergy with other chemicals. For example, combination treatment with EGCG and tamoxifen was synergistically cytotoxic10 and enhanced apoptosis40 in MDA-MB-231 human breast cancer cells and decreased tumor growth in a MCF-7 cell xenograft model.41 Combination treatment of MCF-7 human breast cancer cells with curcumin and genistein also resulted in synergistic growth inhibition,12 while curcumin and gemcitabine decreased pancreatic tumor volume in an orthotopic model.42 Furthermore, the combination of EGCG and curcumin synergistically inhibited the growth of human oral epithelial cells, which the authors postulated was mediated via differential effects on the cell cycle, as EGCG elicited G₁ arrest and curcumin elicited S/G₂M arrest.43 However, these 2 agents have also produced opposing actions, where EGCG promoted differentiation in normal keratinocytes but this effect was antagonized by curcumin.44, 45 This is an important finding but may not apply to cancer models.

Given the contrasting effects produced by these two compounds in various in vitro models, we therefore aimed to determine how EGCG and curcumin would interact in both in vitro and in vivo models of ERα- breast cancer. This type of breast cancer was specifically chosen because EGCG has demonstrated greater cytotoxicity in ERα-breast cancer cells compared to ERα+10 and was also efficacious in MDA-MB-231 xenograft models.8, 36, 46

Material and methods

Chemicals and reagents

Epigallocatechin gallate (EGCG, 99% purity) and curcumin (99% purity) were purchased from the Cayman Chemical Company (Ann Arbor, MI). MDA-MB-231 and T47-D human breast cancer cells were purchased from ATCC (Manassas, VA). Matrigel and the primary antibodies Akt, p-Akt, EGFR, p-EGFR were purchased from BD Biosciences (Auckland, NZ). Minimum essential medium (MEM), Dulbecco's modified eagle's medium nutrient mixture F-12 (DMEM/F-12), potassium chloride, NaHCO₃, NaCl, sulforhodamine B (SRB), trizma hydrochloride, propidium iodide, triton-X 100, trypsin and trypan blue were purchased from Sigma Chemical (St Louis, MO). Acetic acid, disodium hydrogen orthophosphate anhydrous, EDTA, DMSO, potassium dihydrogen orthophosphate, sodium citrate and trichloroacetic acid were purchased from BDH Laboratory supplies (Poole, England). VEGFR-1 antibody was purchased from Abcam (Cambringe, MA). Fetal bovine serum (FBS) was purchased from Life Technologies (Auckland, NZ). All other chemicals were the highest purity commercially available.

Cell culture

Cells were maintained in MEM media supplemented with 10% FBS, 1% antibiotic/antimitotic solution and 0.2% NaHCO₃. Cells were cultured in 75 cm² flasks and incubated in 5% CO₂/95% humidified air at 37°C.

Cytotoxicity assays

Cell cytotoxicity was determined using the sulforhodamine B (SRB) assay as previously described.47 MDA-MB-231 and T47-D human breast cancer cells were seeded in 6-well plates (70,000 cells/well) in 5 ml DMEM/F-12 media supplemented with 5% FBS, 1% antibiotic/antimitotic solution and 0.2% NaHCO₃. After 24 hr, cells were treated with EGCG (20–40 μM), curcumin (2–6 μM) or a combination of the two for 5 days. For time-course studies, MDA-MB-231 cells were plated in 24 well plates at 100,00 cells/well and treated 24 hr later with EGCG (20 μM), curcumin (3 μM), or a combination of the two for 12, 24, 36 or 48 hr. In all cases, vehicle control cells were treated with DMSO (0.1%). Results are expressed as cell number (determined from a standard curve) from 6 independent experiments conducted in duplicate.

Cell cycle distribution by flow cytometry

To determine cell cycle phase distribution, MDA-MB-231 cells were seeded in 24-well culture plates (100,000 cells/well) in 2 ml of DMEM supplemented with 5% FBS, 1% antibiotic/antimitotic solution and 0.2% NaHCO₃. After 24 hr, cells were treated with EGCG (20 μM), curcumin (3 μM) or a combination of the two for 6, 12, 18 or 24 hr. Cell cycle distribution was assessed using propidium iodide staining, as described.48 Samples were analyzed using a fluorescence-activated cell sorter (FACS) FACScalibur (Becton Dickinson, NJ) and data was captured using CellQuest Pro software. Results are expressed as the proportion of cells in G₁/G₀, S or G₂/M-phases from 6 independent experiments performed in duplicate.

Animals and treatment

Female CD1 athymic nude mice (5- to 6-weeks-old) were purchased from Hercus Taieri Resource Unit (Dunedin, NZ). They were maintained at 21–24°C with a 12-hr light/dark cycle in a specifically designed pathogen-free isolation facility and allowed to acclimatize for 1 week before experimentation. The University of Otago Animal Ethics Committee approved all procedures. Mice were inoculated into the right flank with MDA-MB-231 cells (2 × 10⁶/50 μl matrigel), which were left to form palpable tumors. Tumor volume was measured weekly with electronic callipers (L × W × H). When the tumor volume reached 100 mm³, mice (8/group) were randomly assigned to the various treatment groups. Mice were then treated with curcumin (200 mg/kg/day, po), EGCG (25 mg/kg/day, ip), EGCG + curcumin, or vehicle control (5 ml/kg/day) for 10 weeks. Dosing solutions were prepared fresh each day. The curcumin dose was based on previous studies that have shown that 200 mg/kg represents the lower limits of the dose of curcumin achieved when animals are fed 2% dietary curcumin.49, 50 Additionally, we have shown that orally administered curcumin at this dose produces a similar effect on cytochrome P450 (CYP) isozymes as a higher dose (400 mg/kg).51 The dose and route of administration of EGCG was based on our previous work in both male and female mice, which demonstrated that EGCG (25 mg/kg, ip) is well tolerated by the mice and also modulates CYP isoforms.52, 53

Assessment of animal health

Food consumption and body weight were monitored daily throughout the treatment period. Mice were euthanized by CO₂ inhalation 24 hr following the last dose and necropsies were then performed. Blood was collected via the inferior vena cava and placed on ice, while major organs, as well as tumors were excised and weighed. Organ weights are expressed as a percentage of body weight. Plasma was separated and used to determine hepatotoxicity via the plasma marker alanine aminotransferase (ALT) activity using a commercially available kit (Medica Pacifica, Auckland, NZ). Results are expressed as IU/l.

Western blotting of proteins from tumor extracts

Protein extraction from tumor tissue was performed as described,54 with the following modifications. Tumors were homogenized in 0.5 ml 10 mM HEPES buffer (pH 7.9, 1.5 mM KCl, 10 mM MgCl₂, 1 complete mini EDTA-free protease inhibitor cocktail tablet) and incubated on ice for 30 min before centrifugation at 12,000g at 4°C for 10 min. The supernatant was collected and saved. The pellet was resuspended in 20 mM HEPES buffer (pH 7.9, 25% glycerol, 1.5 mM MgCl₂, 420 mM NaCl, 1 complete mini EDTA-free protease inhibitor cocktail tablet) and incubated on ice for 30 min and then centrifuged at 16,000g at 4°C for 30 min. The supernatants from each spin were combined and then fractioned by SDS-PAGE, transferred to nitrocellulose membranes using a semi-dry transblot, blotted with each antibody and detected by amplified color reagent (Bio-Rad, Auckland, NZ). β-actin was used as a loading control and the density of each band was normalized to the β-actin control.

Statistical analysis

All time course data were analyzed using a two-way ANOVA coupled with a Student–Newman–Keuls post hoc test. Tumor volume was analyzed using a repeated measures 2-way ANOVA coupled with a Student–Newman–Keuls post hoc test. For all data in which time was not a factor, the data were analyzed using a one-way ANOVA coupled with a Student–Newman–Keuls post hoc test. p < 0.05 was the minimum requirement for a statistically significant difference.

Results

Cell viability after exposure to combination treatment

Combination treatment of MDA-MB-231 cells with EGCG + curcumin for 5 days produced a greater degree of cytotoxicity than treatment with either compound alone. Specifically, combination treatment with curcumin (3 μM) and EGCG (20 μM) significantly decreased cell number by 77.4% ± 2.5% of control, while EGCG alone decreased cell number by 24.8% ± 2.6% of control (Fig. 1a). To determine if a similar effect would also occur in ERα+ breast cancer cells, T47-D cells were incubated with curcumin (4–6 μM), EGCG (35–40 μM) or a combination of the two. Higher concentrations of both compounds were used, as we have previously shown that both MCF-7 and T47-D cells were more resistant to the cytotoxic effects of EGCG.10 The results show that the combination of EGCG + curcumin was cytotoxic toward T47-D cells (Fig. 1b). However, cell number was not significantly reduced compared to the individual treatments. Additionally, the highest concentrations in combination reduced cell number by only 49%. Therefore, an enhanced effect by the combination treatment was only elicited in MDA-MB-231 cells. Further time course and cell cycle studies were therefore conducted in MDA-MB-231 cells. Results from time course studies showed that synergistic cytotoxicity was first observed 24 hr after treatment of MDA-MB-231 cells with EGCG + curcumin (Fig. 2), where the combination treatment significantly decreased cell number by 49.5% ± 3.8% of control.

Figure 1. Cell number following treatment with EGCG and curcumin. (a) MDA-MB-231 cells were treated with either EGCG (20–25 μM), curcumin (2–3 μM) or a combination of the two for 5 days. Vehicle control cells were treated with DMSO (0.1%). (b) T47-D cells were treated with either EGCG (35–40 μM), curcumin (4–6 μM) or a combination of the two for 5 days. Cell number was determined using the SRB assay. Bars represent the mean ± SEM from 6 independent experiments conducted in duplicate. Data were analyzed using an ANOVA coupled with the Student–Newman–Keuls post hoc test in which p < 0.05 denotes a statistically significant difference. *significantly different to control p < 0.05. **combination significantly different to either compound alone p < 0.05.

Figure 2. Time-course of cytotoxicity following treatment with EGCG and curcumin. MDA-MB-231 cells were treated with either EGCG (20 μM), curcumin (3 μM) or a combination of the two for 12, 24, 36 or 48 hr. Vehicle control cells were treated with DMSO (0.1%). Cell number was determined using the SRB assay. Symbols represent the mean ± SEM from 6 independent experiments conducted in duplicate. Data were analyzed using a two-way ANOVA coupled with the Student–Newman–Keuls post hoc test in which p < 0.05 denotes a statistically significant difference. *significantly different to control p < 0.05. **combination significantly different to either compound alone p < 0.05.

Flow cytometric analysis of MDA-MB-231 cell cycle phase distribution

Analysis of cell cycle progression was conducted in order to examine the mechanisms underlying the cytotoxicity elicited by the combination treatment. FACS analysis was used to determine cell cycle distribution. The largest increase in the proportion of cells in the G₂/M-phase was observed following treatment with 20 μM EGCG + 3 μM curcumin for 12 hr (Fig. 3a). Specifically, the combination treatment increased the proportion of cells in G₂/M-phase by 262.7% ± 15.9% of control. The observed arrest in the G₂/M-phase produced by the combination treatment corresponded with a 39.8% ± 4.0% decrease in the proportion of cells in the G₁/G₀-phase compared to control (Fig. 3b).

Figure 3. Cell cycle progression following treatment with EGCG and curcumin. MDA-MB-231 cells were treated with EGCG (20 μM), curcumin (3 μM) or a combination of the two for 6, 12, 18 or 24 hr. Vehicle control cells were treated with DMSO (0.1%). (a) Proportion of cells in G₂/M phase (b) Proportion of cells in G₁/G₀ phase. Cell cycle progression was determined by PI staining and subsequent flow cytometric assessment of cellular DNA content. Symbols represent the mean ± SEM from four independent experiments conducted in duplicate. Data were analyzed using a two-way ANOVA coupled with the Student–Newman–Keuls post hoc test in which p < 0.05 denotes a statistically significant difference. *significantly different to control p < 0.05. **combination significantly different to either compound alone p < 0.05.

Tumor growth in vivo following combination treatment

To determine the in vivo safety and efficacy of the combination treatment, female athymic nude mice were implanted with MDA-MB-231 cells and treated with EGCG (25 mg/kg), curcumin (200 mg/kg) and a combination of the 2 drugs for a period of 10 weeks. Combination treatment caused a significant reduction in tumor volume over the duration of the study compared to vehicle control (p < 0.003) (Fig. 4). Specifically, after 10 weeks, the mean tumor volume from combination treated mice was 49% smaller than vehicle control mice (tumor volumes of 445 ± 70 and 231 ± 40 mm³, for control and combination treatments, respectively, p < 0.05). The mean tumor volume from EGCG treated mice was 31% smaller, while the curcumin only treated group was not different from vehicle control. Additionally, the mean tumor weight in the combination group was ˜50% smaller than the control group (122 ± 59 mg vs. 241 ± 94 mg, respectively) (Table I). However, this difference was not statistically significant which may be due to the variation in the control group.

Figure 4. Tumor growth following treatment with EGCG and curcumin. Female athymic nude mice were implanted with MDA-MB-231 cells (2 × 10⁶) and treated with either EGCG (25 mg/kg, ip), curcumin (200 mg/kg, po), a combination of the two or vehicle control for 70 days. Tumor volume (LxWxH) was determined using electronic callipers. Symbols represent the mean ± SEM from 8 mice. Data was analyzed using a repeated measures two-way ANOVA coupled with the Student–Newman–Keuls post hoc test in which p < 0.05 denotes a statistically significant difference. *combination treatment significantly different to vehicle control p < 0.03.

Table I. Body Weight Gain, ALT Activity, Organ Weight and Tumor Weight Following Drug Treatment

Treatment (dose)	Body weight gain (g)	ALT activity (IU/L)	Liver weight (% BW)	Kidney weight (% BW)	Spleen weight (% BW)	Tumor weight (mg)
Tumor bearing mice were treated for 70 days. Each value represents the mean ± SEM from 10 mice. Significance was determined with an ANOVA coupled with a Student–Newman–Keuls post hoc test where p < 0.05 was required for a statistically significant difference. * Significantly different compared to the EGCG and vehicle treated groups, p < 0.05.
Vehicle (5 ml/kg)	1.9 ± 0.2	34.7 ± 2.0	5.6 ± 0.1	1.6 ± 0.03	0.48 ± 0.05	241 ± 94
Curcumin (200 mg/kg)	−0.3 ± 0.6*	32.8 ± 5.5	5.9 ± 0.2	1.7 ± 0.05	0.39 ± 0.06	242 ± 88
EGCG (25 mg/kg)	1.3 ± 0.3	28.6 ± 4.6	5.5 ± 0.1	1.7 ± 0.02	0.47 ± 0.05	168 ± 47
Curcumin + EGCG	0.7 ± 0.7	39.1 ± 2.2	5.1 ± 0.4	1.7 ± 0.03	0.43 ± 0.05	122 ± 59

General observations of animal health

While EGCG was well tolerated by the mice, curcumin treatment caused mice to lose 0.3 ± 0.6 g of weight (Table I). However, weight gain in mice treated with curcumin + EGCG was not different from control. All other physiological parameters (i.e., liver, kidney, spleen and uterine weight as well as ALT activity) were normal (Table I) and thus the combination treatment was nontoxic to the mice.

Protein levels of EGFR, Akt and VEGFR-1 in tumors

To determine potential mechanisms for the efficacy of the combination treatment, tumor extracts were analyzed for changes in the expression of proteins involved in both angiogenesis and cell proliferation. The results demonstrated that vascular endothelial growth facotor receptor (VEGFR-1) protein levels in tumors from EGCG + curcumin treated mice were decreased 78% ± 6% compared to control and 42% ± 12% compared to curcumin (Fig. 5). Furthermore, other proteins analyzed (EGFR, p-EGFR, Akt and p-Akt) were all decreased following combination treatment compared to control, but these were not decreased compared to curcumin treatment (Fig. 5). Additionally, both the unphosphorylated and phosphorylated forms were decreased by approximately the same extent following curcumin treatment.

Figure 5. Tumor protein levels of VEGFR-1, EGFR, p-EGFR, Akt and p-Akt following treatment with EGCG and curcumin. (a) Representative Western blots of the various proteins (b) Results of scanning densitometry of Western blots. Bars represent the mean ± SEM from 8 mice. Data was analyzed using an ANOVA coupled with the Student–Newman–Keuls post hoc test in which p < 0.05 denotes a statistically significant difference. *significantly different to vehicle control, p < 0.05, **significantly different to all other treatment groups p < 0.01.

Discussion

Our results have demonstrated that the combination of EGCG + curcumin is synergistically cytotoxic toward MDA-MB-231 human breast cancer cells in vitro and decreases ERα-tumor growth in vivo. However, enhanced cytotoxicity did not occur in T47-D cells. This is not surprising as EGCG is more effective in ERα- breast cancer cells10 and curcumin shows greater activity against ERα+ breast cancer cells when it can act as an antagonist and inhibit estradiol-induced cell proliferation.55 We have also shown that EGCG + curcumin arrested the growth of MDA-MB-231 cells in the G₂/M-phase, an effect that is likely to contribute to the observed cytotoxicity. A number of other studies have also demonstrated that the inhibition of cancer cell growth by either EGCG or curcumin is associated with cell cycle arrest. For example, treatment of DU145 human prostate cancer cells with 20–40 μg/ml EGCG for 24 hr resulted in a dose-dependent arrest in G₀/G₁-phase.14 Additionally, treatment of MDA-MB-468 and HBL100 human breast cancer cells with 20 μM curcumin for 48 hr induced G₂/M-phase arrest. Specifically, the proportion of cells in the G₂/M-phase was increased ˜56% and 40% in MDA-MB-468 and HBL100 cells, respectively.18 Additionally, the two compounds synergistically inhibited the growth of human oral epithelial cells and the authors postulated that this was due to differing effects on the cell cycle, as EGCG elicited G₁ arrest and curcumin elicited S/G₂M arrest.43

One downfall of curcumin is its low bioavailability.56 Therefore, the significant increase in cytotoxicity and cell cycle arrest that occurs following combination therapy may also contribute to an increased efficacy following lower doses of curcumin in vivo. This would be advantageous to patients in chemopreventative trials who are currently taking up to 12 g of curcumin.56 Importantly our in vitro cytotoxicity results correlated with efficacy in a xenograft model of ERα- tumorigenesis. The growth curves in the MDA-MB-231 xenograft model began to diverge after 6 weeks of treatment, where EGCG and EGCG + curcumin both appeared to suppress the rate of tumor growth. Throughout the entire treatment period, the combination treatment exhibited the least amount of growth and produced a mean tumor weight that was 50% smaller than control. This decrease in tumor growth correlated with a 78% decrease in protein levels of VEGFR-1 in the tumor. Thus, the combination treatment may be more effective at inhibiting angiogenesis than either treatment alone, as VEGFR-1 contributes to the regulation of angiogenesis.

Inhibition of angiogenesis appears to be the one dominant mechanism regulating tumor growth, as the modulation of other important cell signaling proteins was not more pronounced following combination treatment compared to curcumin treatment. While there was no further reduction in the active forms of either EGFR or Akt following drug treatment, the similar decrease in both the unphosphorylated and phosphrylated forms indicates that the signaling capacity of the pathways as a whole was reduced ˜75% by curcumin treatment. The large reduction in these proteins elicited by curcumin possibly prevented a further appreciable decrease in protein expression following combination treatment. A similar effect on cell signaling proteins has been reported for curcumin alone57 and the combination of curcumin and gemcitabine.42 Specifically, after 35 days of treatment with curcumin (1 g/kg) + gemcitabine pancreatic tumor volume was decreased (MIA PaCa-2 cells) compared to all other treatments.42 This correlated with a greater decrease in NF-κB-regulated gene products such as, c-myc, ICAM-1, Bcl-xL and survivin in the combination group compared to curcumin or gemcitabine alone. However, NF-κB, COX-2 and VEGF were decreased to the same extent following curcumin and curcumin + gemcitabine.42 A decrease in angiogenesis following curcumin has also been observed in an orthotopic mouse model of ovarian cancer. Specifically, curcumin (500 mg/kg) decreased VEGF and MMP-9 expression as determined by immunohistochemistry and this correlated with a significant decrease in microvessel density.58 Similarly, we observed a large decrease in VEGFR-1 protein levels following EGCG + curcumin and this suggests that the tumor growth was decreased because of lower levels of angiogenic stimulating factors such as VEGF. However, a direct measure of VEGF is needed to confirm this result. Overall, our results are the first to show that EGCG + curcumin is safe and efficacious in a mouse model of ERα- tumorigenesis and mechanistic studies point to a decrease in angiogenesis as a likely mechanism for this effect.

Although, doses of curcumin of up to 12 g have not produced toxic side-effects in clinical trials, the sheer volume of the medication needed to achieve this dose is not well tolerated by the patients.56 Therefore, strategies to increase the bioavailability of curcumin have been examined. Specifically, piperine has been used to decrease the glucronidation of curcumin and thus decrease its excretion,59 however this strategy has not been linked to a decrease in tumor growth. Importantly, our studies demonstrated that an ineffective dose of curcumin could elicit significant tumor suppression when it was combined with EGCG. Additionally, in our study tumor suppression occurred at doses of curcumin that were 2.5- to 5-fold lower than that required for tumor suppression in other animal models of cancer.42, 58 This indicates that, when used in combination with EGCG, lower doses of curcumin would be required in clinical trials. Overall, our findings support the use of a combination of EGCG and curcumin as antitumorigenic therapy in ERα- breast cancer and this warrants further investigation.

Acknowledgements

Funding for this work was obtained from a University of Otago research grant (Dr. RJR) and University of Otago Postgraduate Scholarships (Ms. TJS-E, Ms. MJS, Ms. ECS, and Ms. SPV).